Hepatitis C Virus Infection in Haemodialysis: The ‘No-Isolation’ Policy Should Not Be GeneralizedAgarwal S.K. · Dash S.C. · Gupta S. · Pandey R.M.
All-India Institute of Medical Sciences, New Delhi, India Corresponding Author
Hepatitis C virus (HCV) infection is the most common blood-borne viral infection in haemodialysis. It causes significant morbidity and long-term mortality. Practice of universal precautions has been reported to be sufficient to prevent HCV seroconversion in dialysis units. However, the seroconversion rate remains very high in many dialysis units. A previous study from 1995 to 1998 at our own hospital without isolation showed that nosocomial transmission is the major cause of HCV seroconversion. The present study was therefore conducted with the aim to study the impact of isolation on HCV seroconversion. In this prospective cohort study, with non-probability consecutive sampling, patients with HCV infection were dialysed in an isolated room. In addition, standard universal precautions were practiced. HCV seroconversion rate was compared with the previous study. All patients with end-stage kidney disease (ESKD) admitted to our hospital for renal replacement therapy were included in the present study. At the time of admission, HCV screening was done. All anti-HCV-positive patients were dialysed in an isolated room. While on maintenance haemodialysis, all patients were monthly tested for anti-HCV, aspartate aminotransferase and alanine aminotransferase. Any patient who had HCV seroconversion was transferred to an isolated room for maintenance haemodialysis. Patients with HCV infection were managed by further testing for HCV-RNA and liver biopsy. Every patient who ultimately received renal transplantation at our hospital was also tested for HCV just prior to renal transplantation as well as 3 months after renal transplantation. HCV infection was diagnosed by detecting anti-HCV antibodies using an ELISA-based third-generation diagnostic test kit. Serum bilirubin, aspartate aminotransferase and alanine aminotransferase were assayed using standard laboratory techniques. From March 2003 to February 2006, 1,417 patients were admitted for haemodialysis in our unit. Of these 1,077 (76%) had ESKD. Mean age of patients was 42.47 ± 16.2 (14–94) and 70.39% were males. Patients with ESKD had had more dialysis sessions (10.9 ± 39.5 vs. 4.4 ± 5.95, p = 0.009), more blood transfusions and more pre-existing HCV infections (4.72 vs. 1.5%, p = 0.009) than patients with acute renal failure. Of the ESKD patients, 65.7% were discharged, 9.47% died, 1.85% were shifted to chronic ambulatory peritoneal dialysis and 22.46% patients received renal transplantation. Of the patients who received renal transplantation, HCV seroconversion was detected in 2.75%. In the previous study without isolation practices, the HCV seroconversion rate in transplanted patients was 36.2%. The hazard of HCV seroconversion was 0.97 (95% CI 0.93–1.02, p = 0.2) for each additional dialysis and 1.09 (95% CI 0.88–1.36, p = 0.37) for each additional blood transfusion. The study concludes that isolation of HCV-infected patients during haemodialysis significantly decreases the HCV seroconversion rate.
Copyright © 2009 S. Karger AG, Basel
Hepatitis C virus (HCV) infection is the most common hepatotropic viral infection affecting the patients on haemodialysis. The prevalence in haemodialysis patients ranges from 6% in the United Kingdom to 60% in Poland and Eastern Europe, and 8–36% in North America. In India, various studies showed a prevalence of HCV of 12–45% in haemodialysis [1,2,3,4]. In our own centre, the prevalence of HCV in haemodialysis ranged from 4.3 to 42% depending upon the duration of haemodialysis in our unit, 42% of the patients being at the end of haemodialysis, just before renal transplantation . The major factors affecting the spread of HCV during haemodialysis are nosocomial transmission and transmission through blood and blood component transfusion [6, 7]. With the development of universal screening of blood and blood products for HCV, nosocomial transmission remains the major route of spread in these patients. Isolation of these patients for prevention of spread of HCV during haemodialysis as is being done in the case of hepatitis B virus-infected patients has always been controversial. The same applies to the re-use of dialysers. Some centres isolate and use dedicated machines for dialysing these patients and do not re-use dialysers while others do not isolate and do re-use the dialysers [8,9,10,11,12,13,14,15,16,17]. This controversy is due to the fact that isolation of infected patients increases the cost of dialysis as in most of the dialysis units isolation also means single use of the dialyser. But if isolation prevents the spread of HCV to other patients, in the long run it will not only be much more cost-effective for the community, but it will also decrease morbidity and mortality of new patients. This will also affect the outcome of renal transplantation in these patients. We have shown that chronic liver disease was the cause of death in renal transplantation in the second decade in 25% of patients at our centres  and even during the early post-operative period, it increases the risk of serious infection in renal-transplant patients . With such an increase in the spread of HCV infection in many dialysis units in India, we suggested to isolate HCV-infected patients . This became even more relevant once we found that hospital staff was not strictly following universal precautions .
However, isolation of HCV-infected patients in haemodialysis can be accepted once one can demonstrate that isolation does decrease the spread of HCV infection in haemodialysis settings. There is no published report regarding the impact of isolation on new HCV infections in haemodialysis patients from this part of the world. This type of study is possible in two ways. Either we have two haemodialysis units and isolate in one unit after randomization of patients. In the absence of two haemodialysis units, it is possible to study the impact of isolation in the same unit with the same staff and compare with a historical study done without isolation. We adopted the second approach. Thus, the present study was planned to study the impact of isolation of HCV-infected patients on new HCV infections in a haemodialysis setting and compare these observations with the incidence of HCV infection in the same unit without isolation as a historical control.
Study design: longitudinal experimental study.
Exposure: HCV-free haemodialysis environment by isolating HCV-positive patients.
Outcome variable: development of new HCV infection in haemodialysis patients.
Setting: haemodialysis unit at our hospital.
Sampling: non-probability consecutive sampling.
Inclusion criteria: all patients admitted for renal replacement therapy, i.e. haemodialysis followed by renal transplantation at our hospital.
Exclusion criteria: patients not willing to consent to the study.
Sample size estimation: 271 patients.
For incidence estimation of HCV infection during haemodialysis, to within 10% of the true value with 90% confidence, we required the inclusion of 271 patients . Our Institute’s Ethics Committee approved the study and informed consent was obtained from all the patients.
All the patients with end-stage kidney disease (ESKD) admitted to our hospital for renal replacement therapy were included in the present study. Patients were subjected to haemodialysis twice/3 times a week as needed. At the time of admission for haemodialysis, anti-HCV and HBsAg were tested in each patient. Any patient positive for HCV or HBV was transferred in an isolated room for haemodialysis. All patients were tested monthly for HBsAg, anti-HCV, aspartate aminotransferase and alanine aminotransferase while on maintenance haemodialysis. Any patient who was found to be anti-HCV or HBsAg positive was also transferred to an isolated room for haemodialysis. A patient who had high alanine aminotransferase levels but was negative for anti-HCV and HBsAg was also isolated for haemodialysis unless there was a satisfactory cause for high alanine aminotransferase. This was done as our experience has shown that an unexplained alanine aminotransferase in a haemodialysis setting is mostly due to HBV and HCV viral infection . Patients testing positive for anti-HCV or HBsAg were managed by further testing for HBeAg, HBV-DNA, HCV-RNA and liver biopsy. Every patient was also tested for HBsAg and anti-HCV just prior to renal transplantation. In view of an incubation period of many weeks prior to the detection of anti-HCV after infection, we also did anti-HCV tests on all patients after 3 months following renal transplantation. Any patient found positive for anti-HCV at this post-transplantation time was also considered positive for the purpose of the study. Also, any patient found to be anti-HCV positive within 8 weeks after admission to our unit for dialysis was considered as having a pre-existing HCV infection and not a new infection in our unit.
HCV infection was diagnosed by detecting anti-HCV antibodies using an ELISA test kit (J. Mitra & Co. Ltd., India), which is based on a 3rd-generation ELISA test. Serum bilirubin, alanine aminotransferase and aspartate aminotransferase were tested using standard laboratory techniques. As in our study on the incidence of HCV infection in haemodialysis carried out from 1995 to 1998 in our own unit in a similar set of patients, these patients served as controls because the set-up and all other variables were the same except that those patients were not isolated in spite of being anti-HCV positive. The results of this study were published earlier . It is also important to mention that our haemodialysis unit is primarily oriented to renal transplantation and does not do life-long maintenance haemodialysis. Thus, patients get only a few months of haemodialysis at our hospital. Patients with ESKD who do not have any definite plan for renal transplantation are discharged to attend some other hospital providing life-long maintenance dialysis. These patients are then not followed by us. Patients who are scheduled for renal transplantation at our centre get the maximum duration of dialysis and the longest follow-up both before and after transplantation. As this study will be compared with a historical study as control, the two study periods also need to be clarified in terms of issues which affect the incidence of HCV infection in a haemodialysis unit.
• HCV-positive patients were not isolated.
• HCV-positive patients’ dialyser was re-used following a standard sterilization procedure with formalin as sterilent as for any other patient who had no HCV infection.
• In this study period, anti-HCV was diagnosed using an ELISA-based 3rd-generation test kit of J. Mitra & Brothers Ltd.
• The blood being given to these patients during haemodialysis was also tested for anti-HCV by using the same test kit as part of the study protocol though our hospital blood bank was also regularly testing these blood units for HCV infection.
• HCV-positive patients were isolated in a separate room as this was the intervention in the study protocol.
• HCV-positive patients’ dialyser was not re-used as this was also the intervention in the study protocol.
• In the current period, anti-HCV was also diagnosed using an ELISA-based 3rd-generation test kit of J. Mitra & Brothers Ltd.
• All the blood being given to these patients was tested regularly by the hospital blood bank; we did not test these blood units for HCV infection separately.
• No separate extra measures were taken in addition to the following universal precautions as compared to those applied during the 1995–1998 period.
In both periods, the same dialysis staff was employed in our dialysis unit. Further, all the staff handling patients had also been tested for anti-HCV. Thus practically we can say that except for isolation and absence of re-use of dialysers there was no difference in terms of measures to prevent HCV infection in the current study as compared to the previous study.
From 1st March 2003 to 28th February 2006, 1,417 new patients were admitted for haemodialysis in our unit. Mean age of patients was 42.47 ± 16.2 (14–94) and 70.39% were males. Of all the patients, 1,077 (76%) had chronic kidney disease (CKD, stage V) and 24% acute renal failure (ARF). The percentages of patients having a past history of blood transfusion, a history of jaundice or a family history of jaundice in all patients and separately in patients with CKD and ARF are shown in table 1. None of the patients reported a positive history of intravenous drug use. Of the 1,077 CKD patients admitted for haemodialysis, 708 (65.7%) were discharged, 102 (9.5%) died, 20 (1.9%) were transferred to continuous ambulatory peritoneal dialysis (CAPD), 242 (22.5%) received a renal transplant and 5 (0.5%) were still getting haemodialysis at the time of study completion. The magnitude of HCV infection and the factors that affect HCV seroconversion like duration of dialysis, number of blood transfusions in each group and outcome of both ARF and CKD are shown in table 2. Overall seroconversion of HCV infection in our haemodialysis unit was 0.63% (0.58% in ARF and 0.64% in CKD). There are two important variables other than practice of universal precautions that affect the incidence of HCV infection in the haemodialysis unit, i.e. the number of blood transfusions and duration of dialysis. There is significant variation in these two variables in patients with different outcomes. For the present study, we performed a multivariate analysis using these two variables taking new HCV infection as end point. There was no significant effect of duration of dialysis and number of blood transfusions on HCV seroconversion in our study. The hazard of HCV seroconversion was 0.97 (95% CI 0.93–1.02, p = 0.2) for each additional dialysis and 1.09 (95% CI 0.88–1.36, p = 0.37) for each additional blood transfusion.
|Table 1. Basic demography of patients at the time of enrollment|
|Table 2. Details related to HCV infection in patients with five outcome groups|
Comparison of the present study and the historical control study in terms of outcome of patients with end-stage kidney disease admitted for haemodialysis is shown in table 3. In the present study, a greater percentage of patients was discharged and a smaller percentage of renal transplantations was done as compared to the past study. However, there was no difference in mortality in the two study periods. Except for patients who underwent renal transplantation, most patients were dialysed for a very short period at our hospital. Thus, comparison of HCV seroconversion is meaningful only in patients who remained with us for a reasonable period of time and received a renal transplant. Table 4 shows the comparison of HCV infection and its related factors in the two study periods for the patients who underwent renal transplantation. It shows that mean age, percentage of males and prevalent HCV infection were similar in the two study periods. However, in the present study, patients remained on haemodialysis for a shorter period of time as compared to the past study (p < 0.001). The incidence of HCV infection in the current study with isolation was significantly lower (2.7 vs. 36.2%, p < 0.001) as compared to the previous study without isolation in patients who were transplanted at our hospital. None of the dialysis staff was anti-HCV positive.
|Table 3. Comparison of outcome of CRF patients between two studies|
|Table 4. Comparison of HCV and its factors in two studies in patients who underwent renal transplantation|
There is no doubt that the prevalence of HCV infection in haemodialysis patients is significantly higher than in the general population [6, 24]. Prior to blood screening for HCV, transfusions used to be the main cause of new cases of HCV in dialysis units. However, the correlation between HCV infection and time on dialysis, higher prevalence in haemodialysis than peritoneal dialysis or home haemodialysis, and the highly variable prevalence from unit to unit all suggest that nosocomial transmission remains a major cause for the high prevalence of HCV infection [6, 25, 26]. The occurrence of nosocomial transmission had been confirmed in some of the studies by phylogenetic analysis by identifying clusters of closely related isolates of HCV, both in studies of individual units with high seroconversion rates and multicentric studies [27, 28].
Now the question comes to select an approach to decrease nosocomial transmission. CDC guidelines in 2001 stated: ‘Patients who are anti-HCV positive (or HCV RNA positive) do not have to be isolated from other patients or dialyzed separately on dedicated machines. Furthermore, they can participate in dialyzer reuse programs. Unlike HBV, HCV is not transmitted efficiently through occupational exposures. Thus, reprocessing dialyzers from HCV-positive patients should not place staff members at increased risk for infection’ [ , p. 29]. In the same issue, the guidelines suggested that ‘further studies are needed to clarify the specific factors responsible for transmission of HCV among hemodialysis patients and to evaluate the effect of the current recommendations on prevention and control of HCV infection in this setting’ [ , p. 32]. This clearly means that if the recommendations cannot prevent HCV seroconversion in a dialysis unit, there is scope for changing the recommendation. In the latest guidelines published in 2008 by the Kidney Disease: Improving Global Outcomes foundation in Kidney International , the following points need to be stressed:
(1) The strength of the recommendation that ‘Isolation of HCV-infected patients is not recommended as an alternative to strict infection-control procedures for preventing transmission of blood-borne pathogens’ is a ‘weak’ one.
(2) The Guidelines state that ‘dialyser reuse should be avoided and if this is not avoidable, then strict precautions should be taken’.
(3) These Guidelines also state that ‘If nosocomial transmission continues to occur, despite reinforcement and audit of the precautions required, a local isolation policy may be deemed necessary. HCV-infected patients should be treated by dedicated staff in a separate room, area, or shift (morning, afternoon, or evening), as there is no rationale for using dedicated machines’.
(4) Research recommendations in the guidelines state that ‘If studies such as DOPPS can also capture the hemodialysis unit’s isolation policies, and the routine methods through which the use of strict infection control procedures is reiterated and monitored, the correlation between practice and nosocomial transmission could be strengthened’. This suggests that there is limitation of data on isolation in dialysis units.
All this suggests that there is a trend towards change in the recommendation on HCV during haemodialysis, though these are two different bodies, which are being compared.
In India, almost all haemodialysis units screen for HCV in patients getting dialysis and claim to follow universal precautions. However, in spite of that, recent reports/presentations have shown a high prevalence (17–46%) of HCV in various haemodialysis units [31,32,33] and no publication has reported decreasing HCV seroconversion in the same unit by just following universal precautions. True assessment of HCV in haemodialysis requires information on many aspects like duration of dialysis, test used to diagnose HCV, isolation practices, prevalence of HCV at the time of initiation of dialysis and seroconversion rate. However, from this part of the country, there is a paucity of studies including all this information. A large number of units still have HCV seroconversion rates to the tune of 40–50%. It is not unexpected that these reports are never published. One of the major limitations in isolating these patients is that there is no randomized control trial on the impact of isolation. The trials in this type of issue require that the same hospital should have two haemodialysis units and that the same dialysis staff should dialyse patients in one unit with isolation and in the other unit without isolation. Such type of randomized control trial will be difficulty to perform. The next best approach is that a historical control can be taken when the same staff had done dialysis without isolation and later does it with isolation. This is what has been done in the present study. In the previous study in the same unit with the same staff, we had seroconversions of HCV to the tune of 36.2%, while in the same unit with the same staff but with isolation the seroconversion rate dropped to 2.7%. There is no other study from India which has shown that the HCV seroconversion rate could be decreased only by following universal precautions. We do not want to affirm that isolation should replace universal precaution practices, but isolation of these patients has an important role to play in dialysis units.
Over the last decade, many studies have evaluated the impact of isolation [34,35,36,37,38,39,40], and the salient features of these studies are shown in table 5. The majority of these studies used isolation along with dedicated machines but one study used only a dedicated machine without isolation and could still decrease HCV seroconversion. Two of these studies reported zero seroconversion after isolation practices. Three of these studies analyzed data retrospectively while 4 studies were done prospectively. Our study has also been done prospectively with a historical control. All these studies including our own show that isolating HCV-infected patients plays a definite role in decreasing HCV seroconversion in dialysis units. The policy of ‘no isolation’ often creates medico-legal issues in some places where isolation is being practiced as isolation is costly and is sometimes challenged by the patient’s party based on these guidelines. We suggest that the wordings of the guidelines in this regard should be appropriately amended so that these policies can be individualized in different clinical settings.
|Table 5. Salient features of studies evaluating the impact of isolation|
Limitations of Study
The study has few limitations. The first limitation is that the two studies that are being compared were carried out at different times: 1995–1998 and 2003–2006. One can argue that with time the dialysis staff has become wiser and was following strict universal precautions and that has resulted in a decrease in the seroconversion rate. This possibility cannot be totally ruled out. However, there has been no further training of the staff in terms of universal precaution in our dialysis unit. The second limitation is that a large number of ESKD patients were dialysed for a short time, during which anti-HCV might not have developed, resulting in false negative results. It is possible, as the window period before anti-HCV becomes positive is several months and if patients are dialysed for a short period then testing might have been missed. However, this situation existed in both study periods and would not have affected the comparison. The third limitation is that the duration of dialysis in the present study was significantly shorter than in the past, control study. This was because the waiting period for a living-related renal transplant has decreased in our hospital due to a larger number of transplantations being done every week. However, we have also tested all the patients transplanted at our hospital 3 months after the transplantation. So, the window period was covered by this approach. Any patient found to be positive after 3 months following transplantation has also been considered as positive and classified as a new HCV infection case. The fourth limitation is that this is a single-centre study and for acceptance by all dialysis units, these results need to be confirmed in a multicentre study. However, we feel that comparison of two groups with and without isolation will be very difficult in the same dialysis unit due to obvious reasons. The last limitation is that we could not do HCV-RNA tests in this study, which is useful in seronegative patients who actually are infected. However, this was also the case in the past study and is thus unlikely to affect the comparison.
From the present study it is concluded that HCV is still a common blood-borne infection in the haemodialysis setting. Practicing universal precautions is of utmost importance for preventing HCV seroconversion. However, in units where universal precautions alone are not able to prevent HCV seroconversion, isolation of patients in separate rooms with dedicated staff definitely plays a role in preventing new HCV infections. In view of this, the guidelines in this regard should be appropriately worded so that these policies can be individualized in different clinical settings.
The Indian Council of Medical Research, New Delhi, India, funded this study.
We would like to acknowledge the support of Prof. S.C. Tiwari, Dr. Dipankar Bhowmik and Dr. Sandeep Mahajan and all residents of the Department of Nephrology for their contribution in patient follow-up.
Dr. Sanjay K. Agarwal, Additional Professor
Department of Nephrology, All-India Institute of Medical Sciences
New Delhi 110029 (India)
Tel. +91 981 143 4836, Fax +91 11 2658 8663
Received: May 28, 2008
Accepted: August 21, 2008
Published online: January 16, 2009
Number of Print Pages : 8
Number of Figures : 0, Number of Tables : 5, Number of References : 40
Nephron Clinical Practice
Vol. 111, No. 2, Year 2009 (Cover Date: February 2009)
Journal Editor: El Nahas M. (Sheffield)
ISSN: 1660-2110 (Print), eISSN: 1660-2110 (Online)
For additional information: http://www.karger.com/NEC